Using Aggregated Relational Data to Feasibly Identify Network Structure without Network Data.

Area of Science:

• Social network analysis
• Sociology
• Computational social science

Background:

• Empirical social network research is often limited by the high cost of data collection.
• Traditional methods require extensive resources for gathering detailed network information.

Purpose of the Study:

• To propose an inexpensive and feasible strategy for network elicitation.
• To demonstrate the utility of Aggregated Relational Data (ARD) in social network research.

Main Methods:

• Utilizing Aggregated Relational Data (ARD), which involves responses to questions about link attributes.
• Employing ARD to recover parameters of a network formation model.
• Sampling from a distribution over node- or graph-level statistics using the recovered model.

Main Results:

• The proposed ARD method allows for network parameter recovery.
• The method enables sampling from network statistics distributions.
• Results from two field experiments replicated using ARD showed similar conclusions to traditional network data.

Conclusions:

• Aggregated Relational Data (ARD) offers a cost-effective alternative for social network analysis.
• ARD facilitates empirical network research by reducing data collection expenses.
• The method is robust and yields comparable findings to conventional network data collection approaches.